Artificial neural diagnostics and prognostics: Self-soothing in cognitive systems

Abstract

Self-diagnostics and prognostics in multi-agent processing systems is explored in the context of self-soothing concepts in Neuropsychology. This is one of the first steps to facilitate Systems-Level Thinking in AI. Autonomous or semi-autonomous system must be able to understand, at a system-wide level, how every part of the system is influencing the other parts of the system. This drives the need for complete self-assessment within the AI system. The use of emotional memory and autonomic nervous state recall can be used to provide contextual cognition for system-level diagnostic and prognostics in large-scale systems. The use of an Artificial Cognitive Neural Framework with intelligent information software agents can be utilized to emulate emotional learning to facilitate self-soothing, which equates to self healing in artificial neural systems. This paper describes the architecture and specifications of software agents that are used to provide self-soothing and self-healing constructs for intelligent systems [1].